Studies representative of psychiatric research in diabetes are typically anecdotal, uncontrolled, unreplicated, lacking in clinical utility, and equivocal in their finding. Despite lack of empirical verification, the belief that emotional characteristics affect the course of diabetes remains a common clinical suspicion. The results of two proposed studies will clarify the interplay of psychiatric illness and diabetes. By focusing on the psychiatric syndromes most frequently encountered in diabetics (anxiety and effective disorders), the studies will also furnish data relevant to the everyday management of these patients. Previous investigations have found that (1) 72% of patients with diabetes (both Types 1 and 2) had a lifetime history of at least one criteria-defined psychiatric diagnosis; (2) affective and anxiety disorders occurred at a rate 6-7 times that observed in the general population; (3) the presence of psychiatric illness did not significantly affect longer-term metabolic control as manifested by glycosylated hemoglobin levels; and (4) individual symptoms of poorly controlled diabetes (e.g., fatigue) were often unrelated to blood sugar status and were more frequently referable to the presence of psychiatric illness. These data were derived from patients with relatively severe disease (80% were insulin-managed; 60% had detectable neuropathy) selected from the diabetes registry of a large university teaching hospital. Thus, generalization from these data is hazardous. Study 1 will determine the prevalence of psychiatric illness in a community-based, private-practice sample of diabetics and, in a controlled, systematic follow-up of patients with affective and anxiety disorders, describe the impact of these illnesses on symptom reporting, blood glucose, glycosylated hemoglobin, and insulin requirements. Affective and anxiety disorders in nondiabetics have each been associated with a pattern of hormonal arousal which, in diabetics, could result in metabolic instability; yet, no data are available on the effect of these disorders on diabetic counterregulatory factors. In a similarly controlled fashion, Study 2, conducted with inpatients undergoing annual diabetes evaluations, will determine the separate effects of affective and anxiety disorders on all the Study 1 dependent variables and also on physiolgoic (e.g., heart rate, blood pressure) and counterregulatory factors (e.g., growth hormone, cortisol, catecholamines). In combination, these studies will document more conclusively the prevalence, impact, and interaction of psychiatric illness and diabetes.